Industry News

Industry News

Machine learning AI News covering topics of What we recommend for new users of CloudQuant, How a Star Trek named supercomputer is perfecting Cancer treatments, How Google’s DeepMind is helping the British NHS detect 50 types of eye disease, How AI can improve our daily live by working alongside us as our assistants, Android Pie, the new AI OS for Android mobiles/machines and our latest features to speed up your model development cycle.

AI & Machine Learning News. 13, August 2018

What would you recommend to someone starting out at CloudQuant?

We asked our portfolio managers, product management team and programmers to tell us what they think will help beginners the most here at CloudQuant. We wanted to help everyone in need of a boost to get started on our platform.
Everyone in our company uses the CloudQuant website and coding platform in one way or another. We all use our own application, just like the crowd researchers. When we say that our free backtesting tools are “institutional grade” we really mean it. Every algo we run in our trading and investment strategies is proven in the same backtesting engine as the crowd uses. We rely on the scorecards, the reports, and the simulated trades to ensure that our trading is successful.
Read the full story

How A Supercomputer Named Dr. Crusher Perfected Cancer Treatments For 21 Patients

Sequencing tumor genes and then selecting treatments that target certain mutations is routine in the treatment of several tumor types, including breast and lung cancer. But it’s not common practice for treating hematological malignancies like multiple myeloma, even though oncologists who treat those diseases are well aware that each patient’s cancer has unique characteristics that may make it more or less responsive to certain targeted treatments.
Oncologist Samir Parekh and his colleagues at the Mount Sinai Icahn School of Medicine in New York wanted to change the treatment paradigm in multiple myeloma. So they developed a system that uses an algorithm to match up multiple myeloma patients with drugs that are already on the market to treat other cancers. They tried the resulting recommendations in a group of 21 patients with treatment-resistant multiple myeloma and got positive results in all but five—a response rate that’s so high Parekh believes the idea could be extended to many other cancers.
2018-08-09 00:00:00 Read the full story.

Google’s DeepMind AI can accurately detect 50 types of eye disease just by looking at scans

DeepMind published on Monday the results of joint research with Moorfields Eye Hospital, a renowned centre for treating eye conditions in London, in Nature Medicine. The company said its AI was as accurate as expert clinicians when it came to detecting diseases, such as diabetic eye disease and macular degeneration. It could also recommend the best course of action for patients and suggest which needed urgent care.

  • Google’s artificial intelligence company DeepMind has published “really significant” research showing its algorithm can identify around 50 eye diseases by looking at retinal eye scans.
  • DeepMind said its AI was as good as expert clinicians, and that it could help prevent people from losing their sight.
  • DeepMind has been criticised for its practices around medical data, but cofounder Mustafa Suleyman said all the information in this research project was anonymised.
  • The company plans to hand the technology over for free to NHS hospitals for five years, provided it passes the next phase of research.
  • Google’s artificial intelligence company, DeepMind, has developed an AI which can successfully detect more than 50 types of eye disease just by looking at 3D retinal scans.

Read the full story (Business Insider).
Artificial intelligence as good as human doctors at spotting early signs of blindness – Read the  story from the Telegraph.
DeepMind’s AI can recommend treatment for more than 50 eye diseases with 94% accuracy – Read the story from VentureBeat.
CloudQuant Thoughts… After the recent well-documented failure of Watson in the area of Cancer diagnosis (worse than failure – recommending treatment regimes that could be harmful to the patient) it was good to read these stories of ML successes in medicine including one for their competition – Google’s DeepMind. Again, ML is impressive when used more as a leverage to human ability. Honing in on problem areas and making suggestions seems to be the current ideal step for a lot of ML and AI improvements in workflow.

Ideas on how AI should Improve Our Daily Lives

In most every example of Artificial Intelligence (AI) in business, there is a chat bot or robot or algorithm that attempts to replace a human. We surely will see replacement via AI, but augmentation is also a value approach. In augmentation, a AI process considers various options and provides the human with improved decision options, based on some deeper consideration or at least with more complexity and optionality involved. There are many things we currently do that are doing regularly that are often conducted with limited consideration of options or a poor consideration of complexity. The penalty we experience is in cost, quality of life, and loss of time. Here are some things, that I think, merit an AI process for continued improvement in everyday life.

  • Daily Dynamic Scheduling
  • Achieving Best Pricing on Regular Purchases
  • Deep (and On-going) Searching
  • Cash Management
  • Best Recipe Tonight

2018-08-08 00:00:00 Read the full story.
CloudQuant Thoughts… These suggestions continue the idea that ML and AI are going to be best used, initially, as leverage to human thought and decision-making processes. I tried to find a video of Apple from late 80’s early 90’s demonstrating a Search that they had called “AN AVATAR” who would continue to search for new information on the “World Wide Network” on subjects that you found interesting. Having someone shop for you can save you hours of your precious time for a very small cost. If we throw away 40% of our food every year, maybe eating out is not such a bad idea.

Android Pie: Google Launches New Artificial Intelligence-Powered OS

Google on Tuesday rolled out Android Pie, its highly-anticipated mobile phone operating system. This OS’ major updates focus is on artificial intelligence, which will allow the system to “learn” from the user and customise the Android experience.
With the Android Pie update, the search engine giant has promised that the phone would become more and more tailored and customised with the user’s behaviour and usage. “We’ve built Android 9 to learn from you — and work better for you — the more you use it,” said Google in an official statement.
2018-08-07 07:50:20+00:00 Read the full story.
CloudQuant Thoughts… It is a pleasure to see companies like Google using AI to improve the user interface. I am always reminded of Palm OS’s three clicks rule (developed from the 80/20 rule) that all information should, wherever possible, be reachable within three clicks maximum.
As we continue to develop our CloudQuant web service, we are always trying to think of how to get the most important data to our users in the smallest number of steps. As an example, today we roll out extra features relating to viewing the best and worst trades taken by your backtests. We now sort all backtest days by the absolute PnL and within each date all trades are, by default, sorted by absolute Gross Profit, so now you can quickly access your best and worst days and trades. We have also added keyboard shortcut keys to allow our users to quickly scan through the charts for their trades. See the CloudQuant forum post for more information.

Below the Fold…

Intel AI and 2,500 Xeons bring ‘The Meg’ mega-shark to the big screen

The Meg, a sci-fi film about a giant prehistoric 75 foot-long shark, debuted last week from Warner Bros. and Gravity Pictures. And Intel said today that its artificial intelligence hardware — namely, about 2,500 Xeon Scalable processors — helped bring the creature to life on the big screen.
Scanline VFX used Ziva VFX software and Intel’s Xeon processors to create the creature, known as the megalodon, with lifelike realism (even though these beas…
2018-08-13 00:00:00 Read the full story.

Delayed impact of fair machine learning

“Delayed impact of fair machine learning” won a best paper award at ICML this year. It’s not an easy read (at least it wasn’t for me), but fortunately it’s possible to appreciate the main results without following all of the proof details. The central question is how to ensure fair treatment across demographic groups in a population when using a score-based machine learning model to decide who gets an opportunity (e.g. is offered a loan) and who doesn’t. Most recently we looked at the equal opportunity and equalized odds models.
2018-08-13 00:00:00 Read the full story.

K-Means Clustering, Creating a Simple Trading Rule for Smoother Returns

What is K-means clustering? K-means is an iterative refinement algorithm that attempts to put each data point into a group or cluster. The algorithm starts with initial estimates for the K centroids (centers of the mentioned groups) and continues moving the centroids around the data points until it has minimized the total distance between the data points and their nearest centroid. The user will generally specify K which is the number of centroids (groups). The algorithm can be thought of in two repetitive steps:

  1. Data assignment – Each centroid defines one of the clusters. In this step, each data point is assigned to one of the centroids or clusters. Assignment is typically done based on Euclidean distance.
  2. Centroid Update – Centroids are then recomputed or moved. This is done by taking the mean of all the data points assigned to that centroid’s cluster

2018-08-09 09:04:44+00:00 Read the full story.

Deploying Machine Learning Models is Hard, But It Doesn’t Have to Be

With free, open source tools like Anaconda Distribution, it has never been easier for individual data scientists to analyze data and build machine learning models on their laptops.
So why does deriving actual business value from machine learning remain elusive for many organizations?
Because while it’s easy for data scientists to build powerful models on their laptops with tools like conda and TensorFlow, business value comes from deploying machine learning models into production. Only in production can a deployed model actually serve the business. And unfortunately, the path to production remains difficult for many companies.
2018-08-09 11:32:55-05:00 Read the full story.

Why Is Auto-Keras Gaining Such Popularity?

Auto Keras is the new open-source neural network library built for automated machine learning. This is built upon Keras where one with less knowledge about machine learning can make use of this library to build neural networks. This library makes the job easy with the help of automated search for hyperparameter selection and finding the optimized values. Let us dive into how to get started with it:
The automated machine learning packages are gaining popularity because of their easy-to-implement technique. A person from a non-AI background or one who has very less machine learning knowledge can build and train neural network within a few lines of code. Let us see how it can be installed.
2018-08-13 08:53:15+00:00 Read the full story.

A Machine Learning Approach — Building a Hotel Recommendation Engine

All online travel agencies are scrambling to meet the AI driven personalization standard set by Amazon and Netflix. In addition, the world of online travel has become a highly competitive space where brands try to capture our attention (and wallet) with recommending, comparing, matching and sharing.
In this post, we aim to create the optimal hotel recommendations for Expedia’s users that are searching for a hotel to book. We will model this problem as a multi-class classification problem and build SVM and decision tree in ensemble method to predict which “hotel cluster” the user is likely to book, given his (or her) search details.
2018-08-13 12:42:11.456000+00:00 Read the full story.

Robotic Process Automation (RPA) And Artificial Intelligence Don’t Have Many Things In Common

Researchers across the globe are trying to inculcate technologies such as robotics and AI in their workflows to optimise and automate them. In such processes, Robotic Process Automation (RPA) is one of the most popular terminologies and is often sought-after for handling operational tasks with least manual intervention. In other terms, it is a software that automates low-level tasks.
However, the two terminologies AI & RPA are used interchangeably. It is important to realise that both RPA and AI facilitate a common goal of Intelligent Automation. While RPA is often picturized as a software robot mimicking human actions, AI is the simulation of human intelligence by machines. In this article, we list down 5 major differences between the RPA & AI.
2018-08-13 08:18:46+00:00 Read the full story.

Clustering algorithms for customer segmentation – Towards Data Science

In today’s competitive world, it is crucial to understand customer behavior and categorize customers based on their demography and buying behavior. This is a critical aspect of customer segmentation that allows marketers to better tailor their marketing efforts to various audience subsets in terms of promotional, marketing and product development strategies.
This article demonstrates the concept of segmentation of a customer data set from an e-commerce site using k-means clustering in python. The data set contains the annual income of ~300 customers and their annual spend on an e-commerce site. We will use the k-means clustering algorithm to derive the optimum number of clusters and understand the underlying customer segments based on the data provided.
2018-08-13 04:44:14.557000+00:00 Read the full story.

Ultimate guide to handle Big Datasets for Machine Learning using Dask (in Python)

Have you ever tried working with a large dataset on a 4GB RAM machine? It starts heating up while doing simplest of machine learning tasks? This is a common problem data scientists face when working with restricted computational resources. When I started my data science journey using python, I almost immediately realized that the existing libraries have certain limitations when it comes to handling large datasets. Pandas and Numpy are great libraries but they are not always computationally efficient, especially when there are GBs of data to manipulate. So what can you do to get around this obstacle? This is where Dask weaves its magic! It works with Pandas dataframes and Numpy data structures to help you perform data wrangling and model building using large datasets on not-so-powerful machines. Once you start using Dask, you won’t look back.
In this article, we will look at what Dask is, how it works, and how you can use it for working on large datasets. We will also take up a dataset and put Dask to good use. Let’s begin!
2018-08-09 08:56:19+05:30 Read the full story.

TensorFlow Vs Caffe: Which Machine Learning Framework Should You Opt For?

When it comes to TensorFlow vs Caffe, beginners usually lean towards TensorFlow because of its programmatic approach for creation of networks. TensorFlow has surged ahead in popularity largely because of the large adoption by the academic community. Caffe, on the other hand, has been largely panned for its poor documentation and convoluted code. In this article, we cite the pros and cons of both the frameworks and see how they stack up against each other for the beginners.
2018-08-07 10:14:52+00:00 Read the full story.

Idea of Audit in AI age

Audit has always been a support function for banks that has served as the last line of defence for keeping the organisation safe from external attacks and provide assurance to business & stakeholders about the safely of the business from any malicious attempt to harm the data or assets of the business. With all companies virtually converting into data goldmines, Audit these days transforms itself into an IT function that ensures the ultimate defence of our prized data assets. Imagine Audit as the plumber / janitor who ensures there are no leaks and the supply pipes are safe from any corrosion or any other issues that may have started to infest the system. AI can help with the heavy lifting but…

  • Are these AI libraries plagues with their own risks?
  • Who assessed these libraries for the bias they bring with the data?
  • What type of data set you run through with these libraries?
  • How to manage inherent bias in the data, as it could be Garbage In – Garbage Out case?

2018-08-09 02:32:09 Read the full story.

Artificial intelligence platform launches to tackle financial crime

As financial crime becomes increasingly sophisticated, a new AI tool has been developed to take on fraudsters, minimize the risks of human error and ensure regulatory compliance.
Technology and consulting services company Mindtree has launched an artificial intelligence and machine learning tool designed to help banks improve their ability to detect financial crimes and enhance reconciliation management. The service is powered by a machine-learning platform developed by predictive analytics specialist Tookitaki.
The launch comes as banks and other financial institutions come under pressure from the increasing sophistication of financial crimes worldwide and ever-more complex regulations requiring strict operating and reporting standards. Because manually detecting money laundering, dealing with false alarms and fragmented reconciliation processes can be costly and time-consuming, there is a growing need for financial institutions to automate many of these processes. Additional benefits of automation include the reduction of errors and quicker response times to incidents.
2018-08-07 00:00:00 Read the full story.

Welcome to Kaggle Data Notes : From Russian Tweets to The World Cup That Nearly Came Home

Enjoy these new, intriguing, and overlooked datasets and kernels.

  1. ???? From Hate Speech to Russian Troll Tweets (link)
  2. ???? Data Science Trends on Kaggle (link)
  3. ????Fashion AC-GAN with Keras (link)
  4. ???? (Bio)statistics in R: Part #2 (link)
  5. ???? Segmenting Buildings in Satellite Images (link)
  6. ⚽ World Cup 2018: The One That Nearly Came Home (link)
  7. ???????? The Best “Izakaya” Restaurant in Kyoto (link)
  8. ???? Dataset: Russian Troll Tweets (link)
  9. ???? Dataset: Political Propaganda on Facebook (link)
  10. ???? Dataset: Predict Pakist…

2018-08-09 00:00:00 Read the full story.

Weekly Selection — Aug 10, 2018 – Towards Data Science

  • Essential Math for Data Science – Why and How
  • Why Automated Feature Engineering Will Change the Way You Do Machine Learning
  • Drake – Using Natural Language Processing to understand his lyrics
  • Loading Data from OpenStreetMap with Python and the Overpass API
  • Web Scraping TripAdvisor, Text Mining and Sentiment Analysis for Hotel Reviews
  • Introduction to NLP
  • Getting started with graph analysis in Python with pandas and networkx
  • Data Science A-Z from Zero to Kaggle Kernels Master

2018-08-10 16:03:12.982000+00:00 Read the full story.

A NLP Guide to Text Classification using Conditional Random Fields

The amount of text data being generated in the world is staggering. Google processes more than 40,000 searches EVERY second! According to a Forbes report, every single minute we send 16 million text messages and post 510,00 comments on Facebook. For a layman, it is difficult to even grasp the sheer magnitude of data out there?
News sites and other online media alone generate tons of text content on an hourly basis. Analyzing patterns in that data can become daunting if you don’t have the right tools. Here we will discuss one such approach, using entity recognition, called Conditional Random Fields (CRF). This article explains the concept and python implementation of conditional random fields on a self-annotated dataset. This is a really fun concept and I’m sure you’ll enjoy taking this ride with me!
2018-08-13 08:32:53+05:30 Read the full story.

College Football – Who’s a Fan?

With the college football season about to begin, it is worth a look to see how the teams stack-up, not based on rankings, but based on fan loyalty. Big Data and Data Visualization help us with this and lead to some interesting insights. College football also offers us a great opportunity to look at the Big Data of fandom and the business of college football and even that of universities.
Building on my post that looked at NFL fan allegiance by location, I thought looking at some data visualization of college football allegiance would tell us a bit about who roots for which team and a bit more about how Big Data, Data Science, and data visualization are helping us understand complex problems in life and business.
2018-08-12 00:00:00 Read the full story.

Four key priorities to keep pace in an evolving market place

There’s no escaping the fact we are seeing considerable changes in the way we work. The proliferation of data, rising fraud, digital disruption and changing regulation continue to put pressure on traditional business models, so it’s essential that plans are put in place to successfully move with the times.
Our recent research has identified four key priorities for businesses in this evolving market place, all influenced by technology and consume…
2018-08-09 09:14:24 Read the full story.

Deep Learning in a “Mobile Phone” Environment

Building deep learning models that can execute on mobile runtimes is a very active area of research in the artificial intelligence(AI) space. After all, mobile devices are a significant source of information and host of computations in the modern technology ecosystems. Among the deep learning techniques that have been trying to adapt to the mobile world, none is more relevant than convolutional neural networks(CNNs) given that they are a foundational block to image analysis methods which can unlock the door to many new scenarios for mobile apps.
Google has been among the players leading the charge in the mobile deep learning space with research like Federated Learning or frameworks like TensorFlow Lite. Recently, researchers from the Google Brain team published a paper introducing MNasNet, a new method for designing CNN models that can effectively execute on mobile devices.
2018-08-13 11:53:39.783000+00:00 Read the full story.

Intel Generated $1 Billion In Revenue From Artificial Intelligence Chips In 2017

Sharing their vision at the recently-concluded Data-Centric Innovation Summit, Intel announced that they had made sold $1 billion of artificial intelligence processor chips last year. Intel’s Xeon processor, whose first ancestor was launched 20 years ago. Now, the tech giant announced that more than $1 billion in revenue came from customers running artificial intelligence on Intel Xeon processors in the data centre.
“Our investments in optimising Intel Xeon processors and Intel FPGAs for AI are paying off… In total, since 2014, our performance has improved well over 200 times,” said Navin Shenoy, executive vice president and general manager of the Data Center Group at Intel Corporation.
2018-08-09 13:07:34+00:00 Read the full story.

Comparing Generative Adversarial Network (GAN) to Encoder Decoder Architecture Is Like Comparing Apples To Oranges

Since the deep learning boom has started, numerous researchers have started building many architectures around neural networks. It is often speculated that the neural networks are inspired by neurons and their networks in the brain. Computational algorithms often mimic and copy these biological structures. But there is yet a lot to be discovered about how the brain actually works.
Neuroscience is nowhere close to solving the mystery of the brain. That is why artificial intelligence scientists have to come up with many neural network architectures to solve different tasks. Two of the main families of neural network architecture are encoder-decoder architecture and the Generative Adversarial Network (GAN).
2018-08-09 11:32:49+00:00 Read the full story.

Conversation-as-a-Service: knowledge economies of scale

Chatbots provides the means for empowering everyone, anywhere with better understanding to help them make more informed, contextual decisions. Though much focus has been about the generalist chatbots like Cortana, Alexa, Duplex, Siri and Bixby, a significant paradigm shift will be through specialist chatbots.
Each specialist chatbot will contain synthesised knowledge, enabling the individual to cut through information overload quickly and effectively to reach the best-fit outcome. Specialist chatbots can be made available everywhere for everyone through omnichannel deployment.
2018-08-13 10:49:50 Read the full story.

Kroger inks Ocado grocery delivery deal to battle Amazon threat

What do food, A.I., and flight control have in common? Well, nothing. Unless you’re the U.K. online food retailer Ocado Group. This canny company uses A.I.-programmed flying robots to coordinate shopping basket fills in their depots, which shaves untold hours and operating costs off of the overheads and has seen its stock rocket in value.
U.S. supermarket chain Kroger Co (KR.N) struck a deal with British online grocer Ocado (OCDO.L) to ratchet up its delivery business with the construction of robotically operated warehouses, upping the ante in the battle with Inc (AMZN.O) and sending Ocado shares rocketing. The U.S. grocery industry is dominated by Walmart Inc (WMT.N) and Kroger but has been in upheaval since last summer, when Amazon’s $13.7 billion deal for Whole Foods sent supermarkets scrambling to match the online retailer on home delivery. The Kroger deal announced on Thursday is Ocado’s first in the United States and the British company’s fourth major agreement with retailers in six months.
A new system of grocery order picking is heading to North American shores courtesy of new deal being drawn up by an online grocer across the pond and one of the biggest retailers in the U.S. While this doesn’t sound like an obvious investment choice for domestic stock lovers, there may well be some upside in the venture.
2018-08-13 00:00:00 Read the full story.

This news clip post is produced algorithmically based upon CloudQuant’s list of sites and focus items we find interesting. If you would like to add your blog or website to our search crawler, please email We welcome all contributors.
This news clip and any CloudQuant comment is for information and illustrative purposes only. It is not, and should not be regarded as “investment advice” or as a “recommendation” regarding a course of action. This information is provided with the understanding that CloudQuant is not acting in a fiduciary or advisory capacity under any contract with you, or any applicable law or regulation. You are responsible to make your own independent decision with respect to any course of action based on the content of this post.